A robust harmonization approach for cognitive data from multiple aging and dementia cohorts.

Area of Science:

• Neuroscience
• Cognitive Science
• Biostatistics

Background:

• Standardized cognitive measures for Alzheimer's disease (AD) are lacking, hindering research consistency.
• Varied assessments across studies complicate data pooling and analysis for AD cognitive decline.

Purpose of the Study:

• To develop a harmonized data approach for assessing cognitive decline in Alzheimer's disease.
• To create robust, cross-cohort cognitive composites for AD research.

Main Methods:

• A two-stage approach was used to pool and harmonize international cognitive data.
• Cognitive composites were derived to maximize data utilization from harmonized datasets.

Main Results:

• The derived cognitive composites demonstrated robustness across diverse cohorts.
• These composites showed sensitivity to AD-related cognitive decline, comparable to existing measures like MMSE and PACC.
• A validation cohort confirmed the efficacy of the harmonization approach and composite measures.

Conclusions:

• An accessible pipeline for harmonizing cognitive data is presented.
• This approach facilitates data pooling for developing and validating findings on AD cognitive decline.
• Researchers can leverage this pipeline to integrate their data with large public cohorts for enhanced AD research.